Ten years. Three places. One thread.

End-to-end ownership of the AWS architecture and full lifecycle of a fleet of 6+ enterprise applications supporting a cross-border trucking company (150 employees, 370 tractor-trucks, 500 trailers, 24/7 operation).

• Cost Multiple environments with budgets $350–$3,400/mo; achieved 50–75% reduction via right-sizing, Fargate optimization and storage lifecycle policies.
• Reliability Multi-repo CI/CD with zero-downtime deployments and automated rollback. Designed offline-first sync for mission-critical field operations.
• Security Least-privilege IAM, end-to-end encryption, centralized SSL/TLS multi-domain, per-environment CORS. Migrated apps off direct prod-DB access (28K+ records protected).
• Automation Built a recurring Well-Architected audit system (cost, security, reliability, DB, hygiene) executed by AI agents (Claude Code).

Designed, operated and maintained AWS infrastructure and SaaS products for a portfolio of 5+ clients across hospitality POS, e-commerce, logistics, real estate, fitness and restaurants.

• ECS migration Migrated 3 production microservices (Java/Spring Boot) and 4 batch processes from legacy VPS to ECS Fargate with blue-green zero-downtime deployments.
• Cost ~30% reduction (~$794/mo) via compute right-sizing, RDS IOPS tuning, ECR/CloudWatch lifecycle policies.
• DB performance Resolved deadlock on a 4.2M+ row table (full-scan 4.2M → 34 rows); 85% reduction in connection leaks (137/h → 21/h).
• Replication Designed AWS DMS pipeline with 22 tables in daily CDC for analytics.
• Observability 15+ CloudWatch alarms with category-based runbooks; resolved a 7h HikariPool exhaustion outage.

Teach advanced programming courses at the undergraduate level. Mentor students on professional residency projects and software development.